In eastern Africa, where the HIV-1 infection rate is among the highest in the world, rural adolescent Ugandan women aged 13-19 are six times more likely to be HIV-1 positive that young men of the same age. HIV-1 transmission is enhanced in young women due to female-specific physiological and biological differences and co-factor variables such as ulcerative genital tract diseases. It is widely recognized that these differences in the natural history of HIV infection in young women do not fully account for the 6:1 ratio. An international consensus is emerging that HIV/AIDS research must now focus on the "contextual factors affecting risk-related sexual behaviors". This medical anthropological Ph.D. research will collect ethnographic, survey and epidemiological data to identify those cultural and social factors which put young Ugandan women at most risk of HIV/AIDS. Specifically, the project will examine the relationships between adolescent sexual behaviors, education, work histories and HIV-1 status using Gagnon's theory of sexual scripting. A script is an organized cognitive schema, composed of norms, personal narratives and expectations, and actual negotiated experiences. It is highly structured, knowable and predictable and therefore accessible to the social scientist. The study will purposively sample 260 young men and women between ages 13-19 living in three different areas of western Uganda. A saliva HIV-1 test will determine the total number of HIV-1 positive cases in the sample and is critical for testing research hypotheses, especially the role of school status on HIV risk profiles. In addition, the correlation of HIV status with responses to the demographic and sexual survey will allow a powerful analysis of the actual influences of the socio-economic context on adolescent vulnerability to HIV. To date, no population-based HIV statistics exist for the research district. Eighty of the original sample population of 260 will be randomly invited for individual interviews and focus groups. Open-ended questions and informal discussion will draw out narrative detail on actual sexual experiences. A non-parametric statistical inference model is used to test the research hypotheses (chi-squares) and to produce generalizable data. Qualitative analyses, such as content and matrix analysis, can determine significant patterns in the interview material thereby testing the validity of the sexual script model and contributing to the anthropological study of adolescent sexuality in Africa. Finally, the - research will produce dat for more effective HIV intervention programs in Uganda.